Social capital development on interest-based networks: examining its antecedents, process, and consequences

The proliferation of online networks has shifted a significant portion of human communication to the digital realm. These networks facilitate not only interpersonal communication but also content creation and sharing. A key outcome of online network participation is the accumulation of social capital—the resources embedded within one's social networks. While existing research extensively explores social capital development in relationship-based networks (e.g., Facebook, WeChat), interest-based networks (e.g., Reddit, Pinterest) have received less attention. Interest-based networks differ significantly because users primarily join to exchange information on shared topics, rather than focusing on relationship building. This study addresses this gap by investigating whether and how bridging and bonding social capital accrue on interest-based networks. Understanding this is crucial because successful interest-based networks rely on trust, reciprocity, and attachment—elements underpinned by social capital. However, the development of social capital in these contexts is presumed to be challenging given the focus on content exchange over relationship maintenance. Therefore, this study explores (a) whether social capital develops on interest-based networks and through which activities; (b) how bridging and bonding social capital develop; and (c) whether social capital contributes to the vitality of these networks. The overarching questions are: 1) Does social capital develop on interest-based social networking sites (SNSs), and through what online activities? 2) How do bridging and bonding social capital develop from online activities? 3) Does increased social capital contribute to the vitality of interest-based SNSs? This study examines the impact of two major activity categories on interest-based networks: online interactions (OIs) with content and with humans. It also investigates the interrelationship between bridging and bonding social capital, often treated as parallel pathways in past literature, and demonstrates how both OIs enhance the sense of belonging via these forms of social capital. The findings contribute to a comprehensive conceptual model of social capital development within interest-based networks.

Social capital theory posits that social networks provide access to resources that can be transformed into other forms of capital. It encompasses structural (interconnections within a network), relational (personal relationships), and cognitive (shared values and understanding) dimensions. Research often distinguishes between bridging (loose ties, broad network) and bonding (strong ties, close-knit groups) social capital, each yielding different resources. While prior research has explored social capital in relationship-based networks, showing a positive correlation between active behaviors (e.g., posting, chatting) and social capital development, interest-based networks remain under-researched. This study builds upon existing work by examining the effects of online interactions (OIs) with content (e.g., posting, favoriting) and with humans (e.g., chatting, discussing) on social capital formation. It leverages social penetration theory to posit a sequential relationship between bridging and bonding capital, where bridging acts as a precursor to bonding. Additionally, the study connects social capital to a sense of belonging, an important indicator of network vitality, drawing upon research demonstrating that social capital fosters group attachment and membership.

This study utilized a cross-sectional survey conducted among users of Douban.com, a prominent Chinese interest-based network, using the Wenjuanxing platform. A pilot study (N=73) tested the reliability and validity of the instruments. The main study involved 624 Douban.com users who were recruited through social media platforms and offered a small reward for participation. The sample was predominantly female (58.7%) and aged 18-25 (70.7%). The survey measured: * **Online Interactions (OIs):** OIs with content (rating, marking, reviewing) and OIs with humans (forum discussions, private chatting, liking, commenting) were measured using frequency scales. * **Social Capital:** Bridging and bonding social capital were measured using adapted scales from previous research. Alpha values exceeding 0.7 were obtained for all scales. * **Sense of Belonging:** This was measured using a three-item scale adapted from existing literature (α = 0.900). Two serial mediator model analyses were conducted using SPSS PROCESS (Model 6) to examine the relationships between OIs (with content and humans as independent variables), bridging and bonding social capital (mediators), and sense of belonging (dependent variable). Demographic variables (age, gender, income, education) were included as covariates. Bootstrap confidence intervals (5000 samples) were used to assess indirect effects.

The descriptive statistics showed significant correlations between all variables. The serial mediator analyses yielded the following key findings: **Model 1 (OIs with content as IV):** * OIs with content positively predicted bridging social capital (B = 0.113, p < 0.001) but did not significantly predict bonding social capital directly. * Bridging social capital significantly predicted bonding social capital (B = 0.178, p < 0.001). * Sense of belonging was positively predicted by both bridging (B = 0.550, p < 0.001) and bonding (B = 0.392, p < 0.001) social capital. The indirect effect of OIs with content on sense of belonging via bridging alone was significant (BootCI = [0.025, 0.102]), as was the indirect effect via bridging and bonding (BootCI = [0.002, 0.013]). The indirect effect via bonding alone was not significant. **Model 2 (OIs with humans as IV):** * OIs with humans positively predicted both bridging (B = 0.374, p < 0.001) and bonding (B = 0.392, p < 0.001) social capital. * Bridging social capital significantly predicted bonding social capital (B = 0.154, p = 0.001). * Sense of belonging was positively predicted by OIs with humans (B = 0.149, p < 0.001), bridging (B = 0.550, p < 0.001), and bonding (B = 0.392, p < 0.001) social capital. All indirect effects of OIs with humans on sense of belonging (via bridging alone, bonding alone, and both sequentially) were significant.

This study makes several important contributions. First, it offers an early exploration of social capital development within interest-based networks, demonstrating that relationships and resources can be built even when the primary focus is information exchange. Second, it identifies OIs with content and humans as important precursors to social capital accumulation. Third, it reveals a sequential relationship between bridging and bonding social capital, where bridging acts as a foundation for the development of stronger ties. The findings suggest that in interest-based networks, where users initially lack prior relationships, shallow interactions (bridging) pave the way for deeper connections (bonding). The distinct pathways from OIs with content and humans to sense of belonging are also noteworthy. OIs with humans have a direct effect on sense of belonging, reflecting the immediate social connection fostered by direct communication. In contrast, the indirect effects of OIs with content suggest that fostering relationships (bridging and bonding) is necessary to cultivate a feeling of community. These findings expand our understanding of social capital formation and highlight the significance of both content interaction and interpersonal communication in online communities.

This study provides valuable insights into social capital development on interest-based networks. It highlights the importance of both online interactions with content and humans in building bridging and bonding social capital, which in turn fosters a sense of belonging. This research has implications for the design of interest-based platforms and underscores the potential of these networks to promote social inclusion and cohesion. Future research should utilize longitudinal studies to confirm the sequential relationship between bridging and bonding, explore specific activities within the broad categories of OIs, and examine the generalizability of findings across different interest-based networks and cultures.

The cross-sectional nature of the study limits causal inferences. Future longitudinal research is needed. The reliance on self-reported data may introduce subjectivity. The findings are specific to Douban.com; replication across diverse platforms is warranted. Future research should investigate the impact of specific types of content and interactions, rather than broad categories.